Tissue engineering holds promise for the treatment of cardiovascular disease. For tissue-engineered cardiovascular constructs it is widely believed that a nonthrombogenic endothelial cell (EC) surface will be advantageous. Many studies have therefore attempted to characterize the reactivity of EC with blood, i.e., EC thrombogenicity. Unfortunately, most in vitro studies with EC have employed anticoagulated blood, static or nonphysiologic blood flow conditions, and relatively short blood exposure times. Consequently, the relevance of these results for in vivo applications remains uncertain. Similarly, in the development of tissue-engineered constructs there have been few studies of EC reactivity in vivo, and no systematic studies have been reported that correlate the properties of ECs that are measurable in vitro with in vivo responses. Nonetheless, it is now recognized within the tissue engineering community (and medical device and drug industries in general) that a key impediment to further progress is the lack of predictive animal models that will enable the rational design of constructs, i.e., preclinical models that will enable optimization of construct performance in vivo based on the identification and selective manipulation of key cellular activities in vitro. Our goal is therefore to establish, for blood-contacting EC surfaces, relationships between pro- and anti- hemostatic properties of ECs that can be preconditioned in vitro, and physiologic responses of thrombosis and vascular healing in vivo. To achieve this goal, the hemostatic properties of constructs variably preconditioned in vitro will be correlated with in vivo outcomes, thereby documenting the utility of rational design. The proposed studies will employ baboon endothelial progenitor cells (EPCs) that are readily isolated from whole blood in vitro, and relevant in vivo baboon models of thrombosis and vascular graft intimal hyperplasia. To specifically enhance or inhibit EPC properties that regulate coagulation, platelets, and other hemostatic functions, EPCs will be grown on conventional ePTFE coated with different matrix proteins (collagen or elastin) and subjected to variable hemodynamic preconditioning (static vs. normal shear). Important hemostatic, inflammatory, and mitogenic activities that are orchestrated by EPCs will be assessed using functional, biochemical, molecular, and histologic assays. We will then assess how EPC properties affect: 1) thrombus formation in native, non-anticoagulated blood under physiologic flow conditions (AV shunt model), and 2) healing of EPC-constructs following surgical placement (aorto-iliac graft model). Although the results may suggest new therapeutic strategies, it is not our primary goal to develop an improved vascular graft or to justify the use of EPCs. Rather, by documenting that cellular preconditioning in vitro can predictably improve outcomes in vivo - a principle that can be extended to other cell types and animal models - these results will validate rational design in general, and encourage studies with other constructs and test beds. Ultimately, these technologies will enable the development of biological constructs for use in man. Public Health Relevance Statement (provided by applicant): Tissue-engineered substitutes are being developed for the treatment of cardiovascular disease. A key advancement that must now be realized is to establish in preclinical animal models that predictive relationships exist between important properties of constructs - properties that can be identified, measured, and optimized in vitro - and in vivo physiological responses including thrombosis and vascular healing. The present studies, to be performed using clinically relevant primate models, will establish such correlations for blood-contacting vascular conduits, thereby validating rational design for these devices and encouraging the development of other predictive test beds. These technologies are critical for the development and clinical application of safe and effective biological prostheses.